Translation Exposure and Firm Value, Evidence From Australian ...

More documents

Recommendations

Info

Al-Shboul & Alison 201four-factor regression model is implemented to examine the impact of foreigntranslation adjustment and net income on the firm’s actual stock returns. The authorincludes two control variables in his model: transaction gain/loss and foreign incometax. Based on a prior economic analysis, Louis hypothesised that positive foreigntranslation adjustments of the manufacturing firms in his sample would be associatedwith decreases in firm value. The results of the study were consistent with thishypothesis, prompting the author to voice concerns about an ‘accounting incomecomputation that recognizes a positive translation adjustment as an increase invalue 10 .’Indirect evidence of the value relevance of translation adjustments is provided by theliterature investigating the efficacy of hedging foreign exchange exposure. In asurvey of 160 Swedish firms over the period 1997 – 2001, designed to provideempirical evidence of the determinants of firm’s hedging decisions, Hagelin (2003)found no evidence indicating that translation hedges were used to increase firmvalue. This suggests that, unlike the situation with transaction exposure, hedgingtranslation exposure is likely to be unacceptable to shareholders, unless managersspecifically justify the use of this type of hedging. This suggestion appeared to havesome validity, as Hagelin and Pramborg (2005) found evidence supporting a positiverelationship between hedging translation exposure and the existence of loancovenants, In another study, Hagelin and Pramborg (2004) found that the Swedishfirms in their sample reduced both transaction and translation exposure by the use offinancial hedges.Contending that accounting translation differences are only value relevant to firms ifthey measure economic exposure to exchange rate movements, Pinto (2005) testsfor value relevance using a sample of 109 U.S. multinational firms, with subsidiariesin Mexico and Germany, for the period 1991 to 1996. Rather than using a returnsmodel which investigates a hypothesised reaction of security prices to accountinginformation, and, therefore, incorporates translation differences at face value, Pintouses a ‘levels’ model which searches instead for a correlation between the marketprices of securities and accounting variables. Pinto follows Ohlson’s (1995)procedure by running a regression between the market value (dependent variable) ofequity and both the book value of equity and accounting earnings per share(explanatory variables). Consistent with a suggestion by Lev (1989) that valuationmodels may be better specified when earnings are decomposed, Pinto addsadditional ‘interaction terms’ to his model which serve as proxies for the theoreticalsources of economic exchange rate exposure. These include asset composition,location of foreign investment, and the timing of exchange rate shocks. The mainfinding of Pinto’s study is that foreign currency translation adjustments aresignificantly value relevant when parameter estimates are allowed to vary in crosssection.A major implication is that the results of previous studies, which incorporatedtranslation adjustments at face value, should be treated with caution, as theirmethodologies may have resulted in a masking of the actual sources of exchangerate exposure. The conflicting results of previous studies have prompted us toinvestigate the question of the value relevance of translation adjustments in anAustralian context. To our knowledge, no empirical work using data for Australianmultinational firms, has been published to date201
4. Data and Methodology4.1 DataAl-Shboul & Alison 202Several public sources were accessed to collect the data for the study. Initially, the500 largest firms listed on the Australian Stock Exchange (ASX) were identified foreach fiscal year (the financial year ending on 30 June) for the period, 2000 to 2004.After eliminating delisted and newly listed firms, it was found that 485 firms remainlisted over the total period. Those firms were subsequently restricted by theapplication of specific criteria to select our study sample. Firstly, financial firms wereexcluded. Because the core businesses of these firms were purely concentrated onfinancial services, it was assumed that they were prepared to undertakecomprehensive strategies to hedge all their exposure positions. The elimination offinancial firms reduced the number of eligible firms to 379. Secondly, since the aim ofour study is to examine the relationship between accounting exposure and the valueof Australian multinational firms, only corporations heavily engaged with foreignoperations through foreign subsidiaries are included. Therefore, those firms havingno foreign subsidiaries over the study period were excluded from the sample. Aftermanually surveying the geographical segment reports contained in the notes to thefinancial statements in the annual report of each firm 11 , 181 such firms wereidentified, with a total number of 2,489 foreign subsidiaries located in foreigncountries and geographical regions.Our survey of these subsidiaries, for the 2004 fiscal year, indicated that the vastmajority of the foreign subsidiaries of the 181 sample firms were heavily dispersedacross the following three geographical regions, ranked respectively: Asia-Pacificregion (including the Asia Crisis region) (997 subsidiaries); European Region (734subsidiaries); NAFTA Region (487 subsidiaries) 12 . The numbers of subsidiaries soidentified is consistent with the findings of a survey conducted by the AustralianBureau of Statistics (ABS, 2004), which reported that the vast majority of Australianforeign subsidiaries in 2002/2003 was spread across the three geographical regions:Asia-Pacific (1,586 subsidiaries); Americas (1,111 subsidiaries); Europe (940subsidiaries). Another indication of the direction of the foreign involvement ofAustralian firms, with which our sample data is consistent, is Harcourt (2007).Harcourt reported that the numbers of firms in 2005, with countries in the Asia-Pacificregion as specific export destinations, were, respectively: 17,394 firms (NewZealand), 6,397 firms (Singapore), 4,984 firms (Hong Kong). The number of firmsexporting to the U.S. was 9,253 and to the U.K., 5,486. These findings provideconfirmation that the vast majority of Australian foreign operations are concentratedin the three geographical regions identified in our study. The present study, therefore,examines the effects of accounting exposure on the value a sample of 181 firmsdescribed above, for the period from January 2000 to December 2004.4.2 MethodologyTo examine the relationship between the translated foreign currency gains or lossesand firm value, we use a two-pass market model. Consistent with the assumptionthat market residuals are a function of earnings (Lev, 1989), Martin et al. (1998)incorporated the effect of changes in earnings and shareholders’ equity, resultingfrom accounting translation differences in their model. To detect this effect, they202
Page 2 and 3: Al-Shboul & Alison 193Firstly, the
Page 6 and 7: Al-Shboul & Alison 197group reporti
Page 8 and 9: Al-Shboul & Alison 199to SFAS No. 5
Page 12 and 13: Al-Shboul & Alison 203firstly used
Page 14 and 15: Al-Shboul & Alison 205In addition,
Page 16 and 17: Al-Shboul & Alison 207regions, resp
Page 18 and 19: Al-Shboul & Alison 209Cornell, B. 1
Page 20 and 21: Al-Shboul & Alison 211O'Brien, T. 1

Translation Exposure and Firm Value, Evidence From Australian ...

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?